Top Banner
The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen isotope record of lacustrine diatom silica Svetlana S. Kostrova a , Hanno Meyer b, * , Bernhard Chapligin b , Pavel E. Tarasov c , Elena V. Bezrukova a a A.P. Vinogradov Institute of Geochemistry, Siberian Branch of Russian Academy of Sciences, Favorsky Str. 1a, 664033, Irkutsk, Russia b Alfred Wegener Institute for Polar and Marine Research, Research Unit Potsdam, Telegrafenberg A43,14473, Potsdam, Germany c Institute of Geological Sciences, Palaeontology, Free University Berlin, Malteserstr. 74-100, Building D,12249, Berlin, Germany article info Article history: Available online xxx Keywords: Lake sediments Stable isotopes Biogenic silica Lake Kotokel Palaeoclimate Hydrological changes abstract The last glacial maximum and late glacial environmental and climatic variability in the Baikal region, southern Siberia, Russia has been studied in a sediment sequence from Lake Kotokel, located 2 km east of Lake Baikal, using the oxygen isotope composition of diatom silica (d 18 O diatom ). The purication of diatom frustules involved the process of trimethylsilylation, which has been shown to be suitable for preparation of diatoms for oxygen isotope analysis. The Lake Kotokel d 18 O diatom record presented here spans intervals from about 24.6 to 22.9 ka BP (further referred to as last glacial maximum) and ~16.7e11.5 ka BP (further referred to as late glacial) displaying variations in the oxygen isotope composition between þ26.7 and þ31.2. Overall high d 18 O diatom values of about þ29 to þ31during the two investigated intervals characterize a strongly evaporative lake system in a dry environment and suggest a lower than present lake level due to enhanced evaporation. The Lake Kotokel diatom isotope record is roughly in line with the 60 N summer solar insolation, pointing to a linkage to broader-scale climate change, but displays weaker reaction to short-term climatic oscillations, i.e. Bølling-Allerød or Younger Dryas. The climate warming at ~14.3 ka BP is marked by negative spikes in the d 18 O diatom record due to isotopically low melt water input from the mountainous hinterland. © 2014 Elsevier Ltd and INQUA. All rights reserved. 1. Introduction Lacustrine sediments are unique carriers of information on past environmental conditions. They record the dynamics of biological productivity and physicochemical processes proceeding through entire lake life. Promising for studies are small shallow lakes, where a high velocity of sedimentation permits palaeoclimatic parameters to be recorded with annual to decadal resolution. A minimum of post-sedimentation changes in bottom sediments ensures accuracy of the acquired knowledge on past environmental and climate conditions at regional and global scale (Last and Smol, 2001; Peck et al., 2002; Subetto, 2009; Swann et al., 2010; Hernandez et al., 2013; Leng et al., 2013; Li et al., 2013). In the past decade researchers have paid special attention to investigating past climate changes and factors leading to switches between glacial and interglacial conditions. Numerous studies have shown the transi- tion from cold and dry glacial to warm and wet interglacial climate led to major environmental alterations in the terrestrial part of lake catchments and in aquatic ecosystems (Morley et al., 2005; Stebich et al., 2009; Müller et al., 2010; Hernandez et al., 2011; Cook et al., 2012; Zhao et al., 2013). The millennial- to century-scale climatic events, such as the last glacial maximum (LGM), the Meiendorf (MD), Bølling (BO) and Allerød (AL) warming phases, the Oldest Dryas (OstD), Older Dryas (OD) and Younger Dryas (YD) cold events were exposed by high- resolution and accurately dated proxy records from terrestrial ar- chives (Wang et al., 2001; Yuan et al., 2004; Svensson et al., 2008) and lake sediments (Litt and Stebich, 1999; Yu and Eicher, 2001; Nakagawa et al., 2005; Diefendorf et al., 2006; Stebich et al., 2009; Zhao et al., 2013), providing important information about the rate, amplitude and driving mechanisms of these climatic events. * Corresponding author. E-mail addresses: [email protected] (S.S. Kostrova), [email protected], [email protected] (H. Meyer), [email protected] (B. Chapligin), [email protected] (P.E. Tarasov), [email protected] (E.V. Bezrukova). Contents lists available at ScienceDirect Quaternary International journal homepage: www.elsevier.com/locate/quaint http://dx.doi.org/10.1016/j.quaint.2014.07.034 1040-6182/© 2014 Elsevier Ltd and INQUA. All rights reserved. Quaternary International xxx (2014) 1e12 Please cite this article in press as: Kostrova, S.S., et al., The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen isotope record of lacustrine diatom silica, Quaternary International (2014), http://dx.doi.org/10.1016/ j.quaint.2014.07.034
12

The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

Aug 24, 2019

Download

Documents

vodien
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

lable at ScienceDirect

Quaternary International xxx (2014) 1e12

Contents lists avai

Quaternary International

journal homepage: www.elsevier .com/locate/quaint

The last glacial maximum and late glacial environmental and climatedynamics in the Baikal region inferred from an oxygen isotope recordof lacustrine diatom silica

Svetlana S. Kostrova a, Hanno Meyer b, *, Bernhard Chapligin b, Pavel E. Tarasov c,Elena V. Bezrukova a

a A.P. Vinogradov Institute of Geochemistry, Siberian Branch of Russian Academy of Sciences, Favorsky Str. 1a, 664033, Irkutsk, Russiab Alfred Wegener Institute for Polar and Marine Research, Research Unit Potsdam, Telegrafenberg A43, 14473, Potsdam, Germanyc Institute of Geological Sciences, Palaeontology, Free University Berlin, Malteserstr. 74-100, Building D, 12249, Berlin, Germany

a r t i c l e i n f o

Article history:Available online xxx

Keywords:Lake sedimentsStable isotopesBiogenic silicaLake KotokelPalaeoclimateHydrological changes

* Corresponding author.E-mail addresses: [email protected] (S.S. Kost

[email protected] (H. Meyer), [email protected] (P.E. Tarasov), bezrukova@

http://dx.doi.org/10.1016/j.quaint.2014.07.0341040-6182/© 2014 Elsevier Ltd and INQUA. All rights

Please cite this article in press as: Kostrova,Baikal region inferred from an oxygen isotoj.quaint.2014.07.034

a b s t r a c t

The last glacial maximum and late glacial environmental and climatic variability in the Baikal region,southern Siberia, Russia has been studied in a sediment sequence from Lake Kotokel, located 2 km east ofLake Baikal, using the oxygen isotope composition of diatom silica (d18Odiatom). The purification of diatomfrustules involved the process of trimethylsilylation, which has been shown to be suitable for preparationof diatoms for oxygen isotope analysis. The Lake Kotokel d18Odiatom record presented here spans intervalsfrom about 24.6 to 22.9 ka BP (further referred to as ‘last glacial maximum’) and ~16.7e11.5 ka BP(further referred to as ‘late glacial’) displaying variations in the oxygen isotope compositionbetween þ26.7 and þ31.2‰. Overall high d18Odiatom values of about þ29 to þ31‰ during the twoinvestigated intervals characterize a strongly evaporative lake system in a dry environment and suggest alower than present lake level due to enhanced evaporation. The Lake Kotokel diatom isotope record isroughly in line with the 60� N summer solar insolation, pointing to a linkage to broader-scale climatechange, but displays weaker reaction to short-term climatic oscillations, i.e. Bølling-Allerød or YoungerDryas. The climate warming at ~14.3 ka BP is marked by negative spikes in the d18Odiatom record due toisotopically low melt water input from the mountainous hinterland.

© 2014 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction

Lacustrine sediments are unique carriers of information on pastenvironmental conditions. They record the dynamics of biologicalproductivity and physicochemical processes proceeding throughentire lake life. Promising for studies are small shallow lakes, wherea high velocity of sedimentation permits palaeoclimatic parametersto be recorded with annual to decadal resolution. A minimum ofpost-sedimentation changes in bottom sediments ensures accuracyof the acquired knowledge on past environmental and climateconditions at regional and global scale (Last and Smol, 2001; Pecket al., 2002; Subetto, 2009; Swann et al., 2010; Hernandez et al.,2013; Leng et al., 2013; Li et al., 2013). In the past decade

rova), [email protected],[email protected] (B. Chapligin),igc.irk.ru (E.V. Bezrukova).

reserved.

S.S., et al., The last glacial mpe record of lacustrine diato

researchers have paid special attention to investigating past climatechanges and factors leading to switches between glacial andinterglacial conditions. Numerous studies have shown the transi-tion from cold and dry glacial to warm and wet interglacial climateled to major environmental alterations in the terrestrial part of lakecatchments and in aquatic ecosystems (Morley et al., 2005; Stebichet al., 2009; Müller et al., 2010; Hernandez et al., 2011; Cook et al.,2012; Zhao et al., 2013).

The millennial- to century-scale climatic events, such as the lastglacial maximum (LGM), the Meiendorf (MD), Bølling (BO) andAllerød (AL) warming phases, the Oldest Dryas (OstD), Older Dryas(OD) and Younger Dryas (YD) cold events were exposed by high-resolution and accurately dated proxy records from terrestrial ar-chives (Wang et al., 2001; Yuan et al., 2004; Svensson et al., 2008)and lake sediments (Litt and Stebich, 1999; Yu and Eicher, 2001;Nakagawa et al., 2005; Diefendorf et al., 2006; Stebich et al.,2009; Zhao et al., 2013), providing important information aboutthe rate, amplitude and driving mechanisms of these climaticevents.

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 2: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e122

The Lake Baikal region in the extremely continental part ofEurasia (Fig. 1a) located far from oceanic influences is one of theworld's most sensitive regions to abrupt climate shifts and multi-millennial variability (Prokopenko et al., 2001; Mackay, 2007;Tarasov et al., 2007; Kuzmin et al., 2009; Sklyarov et al., 2010;Shichi et al., 2013). To date, the last and late glacial climate andenvironment development history in the region was primarilyreconstructed based on multi-proxy data from Lake Baikal(Prokopenko et al., 2001; Karabanov et al., 2004; Bo€es et al., 2005;Demske et al., 2005; Morley et al., 2005; Mackay, 2007; Kuzminet al., 2009; Mackay et al., 2011) and using pollen and diatom re-cords from the relatively small (about 69 km2) and shallow (about4 m mean water depth) Lake Kotokel (Zhang et al., 2013) locatednear Lake Baikal (Fig. 1) (e.g. Bezrukova et al., 2008, 2010, 2011;

Fig. 1. Schematic maps showing (a) location of the study area in Eurasia; (b) the area arountriangle) and water sampling sites (black rhombus) used in this study; and (c) the Baikal r

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

Shichi et al., 2009; Tarasov et al., 2009; Müller et al., 2013). Theobtained high-resolution sedimentary records from Lake Kotokel(Müller et al., 2013) revealed the cold and dry LGM (between 26.5and 19 ka BP; here and throughout the entire article calendar agesare consistently used) with highest scores for steppe biome, minorwoody coverage, very low diatom concentrations and low lakelevel. A slight climate amelioration is reconstructed at~24e22 ka BP. A gradual increase in tree/shrub pollen percentagesand increase in diatoms is noticeable at ~17.0e11.65 ka BP(Bezrukova et al., 2010). The deepening of the lake and increase inthe woody coverage to 20e30% at ~14.5e14.0 and ~13.3e12.8 ka BPgave clear evidence for the MD and AL interstadials. The YD coolingin the Baikal region at ~12.7e11.65 ka BP is well recognizable in theLake Kotokel records by an increase in tundra biome scores and

d Lake Kotokel (52�500N, 108�100E, 458 m a.s.l.) with location of the KTK2 core (blackegion with recent diatom sampling site position (black pentagon).

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 3: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e12 3

pronounced changes in the diatom species composition (Bezrukovaet al., 2010) and in Lake Baikal through variations in diatomabundance and diatom d18O values, organic matter and d13C valuesas well as by a lowering of grey scale values (Prokopenko andWilliams, 2004; Bo€es et al., 2005; Kuzmin et al., 2009; Mackayet al., 2011). The variations in each proxy through the last glacia-tion are considered to be due to variability in biological activity andsedimentation rates.

A number of qualitative climatic and environment re-constructions published to date demonstrate a complicated andpartly controversial picture of climate changes during the LGM andlate glacial in the Baikal region (Bo€es et al., 2005; Demske et al.,2005; Bezrukova et al., 2010; Mackay et al., 2011). Therefore, oxy-gen isotope data from small lakes in the region could be helpful inresolving some of the problematic questions, as recently demon-strated by the Holocene oxygen isotope record of diatoms fromLake Kotokel (Kostrova et al., 2013a, b).

Because the oxygen isotope analysis of diatom silica requireshighly pure samples, various chemical and physical procedures needto be applied to separate diatom frustules from sediment material(Morley et al., 2004; Leng and Barker, 2006; Swann and Leng, 2009;Chapligin et al., 2012; and references therein). The use of differentpurification approaches is often unavoidable and dictated by theavailability of certain chemicals and laboratory equipment.Hydrogen peroxide (H2O2) and various acids (HNO3, HClO4, HCl) aregenerally used for organic and carbonate removal. Clay, silts andother remaining contaminants are then separated by sieving at asize fraction, which needs to be adjusted according to the size ofdiatom frustules and contaminants in the sediment assemblage.Heavy liquid separation (HLS) with sodium polytungstate (SPT;3Na2WO4$9WO3$H2O) is also used (Morley et al., 2004; Chapliginet al., 2012). Within this study the clean-up procedure involvedthe process of trimethylsilylation (Laufer and Roy, 1972; Kashutinaet al., 1975; Iler, 1979). Even though the approach was employedearlier by Kalmychkov et al. (2005, 2007) and Kostrova et al. (2013a,b), the question whether the trimethylsilylation reaction (TMSR) byusing trimethylchlorosilane (TMCS; (CH3)3SiCl) influences the finaloxygen isotope composition of diatoms is still under debate and istherefore additionally investigated within this study. This informa-tion is needed to assess whether diatom oxygen isotopes preparedwith this technique are reliable proxies for palaeoclimate recon-struction. We furthermore aim to present the recently obtainedoxygen isotope record of diatoms extracted from Lake Kotokelsediments accumulated ~24.6e22.9 ka BP and ~16.7e11.5 ka BP.Together with the existing reconstruction derived from pollen anddiatom records from the same sediment core and the other pub-lished data, our new record is then used to discuss the environ-mental dynamics and climate history of the region for the last glacialmaximum and the late glacial.

2. Regional setting

2.1. Site location, morphology and hydrology of the lake

Lake Kotokel (458 m a.s.l.; Fig. 1), situated between the Kika andTurka rivers in south-eastern Siberia, is 15 km long and about 6 kmwidewith a catchment area of approximately 187 km2 (Zhang et al.,2013) and a relatively short water residence time of about 7 years(Shichi et al., 2009). The lake basin (52�450e52�520 N;108�040e108�120 E) of Cenozoic age (Florensov, 1960) is separatedfrom Lake Baikal by a low-elevation mountain ridge (up to729 m a.s.l.). Both the southern and northern parts of the lake aresimilar in basin morphology, comprising an almost flat lake bottom(Zhang et al., 2013). The lake has an outflow to Lake Baikal via theIstokeKotochikeTurka river-system and depending on the

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

precipitation amount in the KotokeleKotochik catchments, theRiver Istok might seasonally flow into Lake Kotokel (Kuz'mich,1988; Kostrova et al., 2013a). However, there is no evidence thatthe Lake Baikal water (and diatoms) has penetrated to Lake Kotokelduring the last 50 ka (Shichi et al., 2009; Bezrukova et al., 2010;Zhang et al., 2013). The lake is well-mixed, the average watertemperature from May to October is about 18 �C and it is generallyice-covered fromNovember to earlyMay (Kostrova et al., 2013a andreferences therein).

2.2. Modern climate

The Baikal region, including the study area, is characterized by acontinental climate with long, cold and relatively dry winters andshort, moderately warm and wet summers. Mean daily air tem-peratures around Lake Kotokel in July are þ15.4 �C, falling toabout�19.5 �C in January (Galaziy, 1993). Westerly winds prevail inthe region throughout the year (Lydolph, 1977). Annual precipita-tion sums reach approximately 400 mm (Galaziy, 1993). Almost ahalf of it falls in July and August during increasing south-easterncyclonic activity along the Mongolian branch of the Polar front,whereas between late autumn and early spring, when cold andsunny weather associated with the Siberian High centred overeastern Siberia and Mongolia predominates, precipitation isgenerally low (Lydolph, 1977; Kurita et al., 2004; Bezrukova et al.,2008; Tarasov et al., 2009).

3. Material and methods

3.1. Sampling, core lithology and age determination

In August 2005 a 12.53 m long sediment core (KTK2; 52�470N,108�070E; water depth of about 3.5 m) was retrieved using aLivingston-type piston corer from the southern part of Lake Kotokel(Fig. 1b) and subsequently studied for lithology, pollen and diatoms(Shichi et al., 2009; Bezrukova et al., 2010). In this study, a530 cm-long section of Lake Kotokel sediments (1030e500 cm ofthe core depth) was chosen for diatom oxygen isotope investiga-tion. A recent lithological study on the KTK2 core (Bezrukova et al.,2010) reported that the selected part of the KTK2 core (Fig. 2a)consists of grey to dark-grey silty clay (1030e1010 cm), laminatedgrey silty clay (1010e740 cm), grey-blackish slightly laminated siltyclay (740e660 cm) and soft brownish-black gyttja (660e500 cm)deposited between ~31.9 and ~11.5 ka BP. The details of the KTK2core chronology was established on the basis of eleven calibratedAMS radiocarbon dates and the age-depth model for the studiedsection are presented in Bezrukova et al., 2010 and in Fig. 2a.

For the investigation of the TMSR purification as an alternative toHLS, a samplewasusedwhere fossil diatomswere separated fromtheterrigenous fraction of Lake Baikal sediments (sample “PL”). A secondsample used for this test (sample “B1”) comprised recent diatomscollected in the central basin of Lake Baikal (53�180N, 108�230E; seeFig. 1c for sampling locations) with net-trap equipment in June 2013.Both samples were dominated by Aulacoseira baicalensis species (upto 98.5%). Additionally, the PL-sample contained Aulacoseira islandicaand Aulacoseira subarctica, whereas the B1-sample contained Cyclo-tella baicalensis and Synedra acus. To test the effect of TMSR on di-atoms both samples were divided in two equal parts and exposed toSPTand TMCS. The oxygen isotope compositionwasmeasuredbeforeand after the respective purification steps were taken.

3.2. Diatom record

The diatom record from the KTK2 core of Lake Kotokel presentedby Bezrukova et al. (2010) reveals a total of 143 diatom taxa through

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 4: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

Fig. 2. Summary of the sedimentary, diatom and isotopic records from Lake Kotokel discussed in this study, including (a) lithology column and calibrated radiocarbon dates; (b)simplified diatom diagram and (c) total diatom concentration (modified after Bezrukova et al., 2010); (d) measured d18O (grey dashed line) and contamination-corrected d18O valuesof diatoms (bold black line); highly contaminated samples (encircled) were excluded; (e) contamination assessment: SiO2 and Al2O3 concentrations in the KTK2 core diatomsamples analyzed by EDS.

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e124

the whole sediment sequence accumulated since about 47 ka BP.The diatom record for the part of the core selected for this study ispresented in Fig. 2b. The intervals from ~31.9 to ~24.7 ka BP(1030e895 cm) and from ~22.0 to ~17.0 ka BP (820e720 cm) revealextremely low diatom concentrations (Fig. 2c) and have beenexcluded from further analysis. The interval ~24.7e21.9 ka BP isdominated by the small benthic Staurosirella pinnata agg. complex(up to 82%; Fig. 2b). However, the total diatom concentration re-mains low, i.e. between 0.3 and 10.3 � 106 valves g�1 (Fig. 2c). Aprogressive increase in the abundance of planktonic Aulacoseiragranulata (up to 91%), Aulacoseira ambigua (up to 34%), Cyclotellaocellata (up to 35%) diatoms is accompanied by a pronounceddecrease in benthic species Pseudostaurosira brevistriata, S. pinnataagg. and Ophephora martyi in the interval between 720 and543 cm KTK2 core depth dated to ~17.0e12.7 ka BP (Fig. 2aand b). The total diatom concentration reaches its maximum of156.3 � 106 valves g�1 at ~14.3 ka BP and decreases to19.1 � 106 valves g�1 towards ~12.7 ka BP (Fig. 2c). The diatomassemblages between ~12.7 and 11.5 ka BP (543e500 cm) aredominated by A. granulata (up to 77%), Ellerbeckia arenaria var.arenaria (up to 84%) and E. arenaria var. teres (up to 15%; Fig. 2b).

3.3. Water sampling and stable water isotope analysis

Surface water from Lake Kotokel and water from the riversconnected to the lake were sampled during May and July 2011, inMarch, September and November 2012 as well as in July and August2013. The water sampling site locations are presented in Fig. 1b.Atmospheric precipitation samples were collected during the timeperiod from June 2011 to October 2013 in Irkutsk, about 270 kmwest of Lake Kotokel. After sampling, all specimens were stored

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

cool in airtight bottles prior to isotope analyses. A total of 178 watersamples were analyzed for hydrogen and oxygen isotopes with aFinnigan MAT Delta-S mass spectrometer at the Isotope Laboratoryof the AlfredWegener Institute for Polar andMarine Research (AWIPotsdam, Germany) using equilibration techniques. Stable isotopedata (d18O, dD and d excess), including minimum, mean andmaximum values, standard deviations (SD) for the analyzed sam-ples, as well as water and air temperatures are summarized inTable 1. Data are given as per mil difference to V-SMOW, with in-ternal 1s errors of better than ±0.8‰ and ±0.1‰ for dD and d18O,respectively (Meyer et al., 2000).

3.4. Sampling preparation and purity estimation

In total, 63 samples with a 4-cm step (an average temporalresolution of about 110 years) from diatoms-containing intervals ofthe KTK2 core dated to the LGM and late glacial (Bezrukova et al.,2010) were prepared for diatom oxygen isotope analysisfollowing a separation and cleaning method first applied byKalmychkov et al. (2005) at the Vinogradov Institute of Geochem-istry SB RAS (VIG SB RAS, Irkutsk) and described in Kostrova et al.(2013a). This procedure includes H2O2 and nitric/perchloric acids(HNO3:HClO4) mixture treatment and sieving with 5 mm mesh sizeto remove organic matter and clay particles, respectively, as well asa modification of diatom surfaces using TMSR (Laufer and Roy,1972; Kashutina et al., 1975; Iler, 1979):

≡SieOH þ (СН3)3SieCl / ≡ SieOeSi(СН3)3 þ HCl(gas)[ (1)

TMSR is conducted in a non-polar environment, i.e. n-hexane(C6H14) or chloroform (CHCl3) in order to avoid any additional

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 5: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

Table

1Su

mmaryof

stab

leisotop

edata( d

18O,d

Dan

ddex

cess),includingminim

um,m

eanan

dmax

imum

values,standarddev

iation

s(SD)as

wella

sslop

esan

dintercep

tsfrom

thed1

8OedD

diagram

forthean

alyz

edsamplesof

Lake

Kotok

elsu

rfacewater,w

ater

from

rive

rsco

nnectedto

Lake

Kotok

elan

dprecipitation

water

from

Irku

tsk.

Sample

type

No.

total

T[�C]

Mea

nd1

8O

[‰]

Min

d18O

[‰]

Mea

nd1

8O

[‰]

Max

d18O

[‰]

SDdD

[‰]

Min

dD[‰

]Mea

ndD

[‰]

Max

dD[‰

]SD

d[‰

]Min

d[‰

]Mea

nd[‰

]Max

d[‰

]SD

Slop

eIntercep

tR2

Surfacewater,L

ake

Kotok

el27

T water,þ1

8.7

�13.7

�12.2

�10.8

0.6

�114

.7�1

07.2

�101

.23.3

�15.0

�9.9

�4.7

2.1

5.0

�45.9

0.92

Inflow

rive

rs,L

ake

Kotok

el10

T water,þ1

2.4

�22.0

�19.7

�16.9

1.3

�161

.4�1

44.5

�126

.29.1

þ9.1

þ12.7

þ14.5

1.5

6.5

�7.0

0.99

Istokrive

r,20

113

T water,þ2

2.8

�20.8

�20.4

�20.2

0.3

�153

.5�1

51.0

�149

.52.1

þ12.0

þ12.3

þ12.7

0.3

6.9

�10.5

1.00

Istokrive

r,20

12e20

134

T water,þ1

1.7

�12.9

�12.1

�11.1

0.8

�112

.4�1

08.0

�102

.24.3

�13.5

�10.9

�8.6

2.2

5.5

�40.8

0.99

Kotoc

hik

rive

r,20

112

T water,þ1

6.5

�21.8

�21.6

�21.4

0.3

�158

.8�1

57.5

�156

.31.8

þ14.8

þ15.0

þ15.3

0.4

1.0

�0.04

1.00

Kotoc

hik

rive

r,20

12e20

136

T water,þ1

0.3

�18.8

�17.5

�14.2

1.9

�144

.0�1

36.6

�118

.510

.1�5

.1þ3

.3þ6

.14.9

5.4

�42.1

0.99

Rainwater,Irkutsk

76T a

ir,þ1

2.4

�24.8

�11.8

�4.7

3.9

�113

�95.0

�47.1

26.9

�19.0

þ0.1

þ15.1

8.1

6.6

�16.8

0.93

Snow

water,Irkutsk

47T a

ir,�1

0.9

�39.5

�27.7

�12.0

5.8

�318

.5�2

10.5

�79.6

48.5

�10.3

þ11.1

þ32.4

8.7

8.2

16.5

0.97

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e12 5

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

oxygen and facilitate the removal of by-products (Kalmychkovet al., 2005). During this process protons of surface silanol groups(≡SieOH) are substituted by non-polar radicals, resulting in di-atoms becoming hydrophobic. Hence, the process causes diatommaterial to float on top of a water column, while all other compo-nents sink. Depending on the degree of sample purity an additionalHLS was performed.

The purity of the obtained diatom preparations was estimatedby the method of Chapligin et al. (2012) using Scanning ElectronMicroscopy (SEM) and Energy-Dispersive X-ray Spectroscopy (EDS)with a ZEISS ULTRA 55 Plus Schottky-type field emission scanningelectron microscope equipped with an energy-dispersive systemand a silicon drift detector (UltraDry SDD; Thermo Fisher Scientific)at the German Research Centre for Geoscience (GFZ), Potsdam. Thequantitative analysis was performed using the standardless pro-cedure (3e5 repetitions, an acceleration voltage of 20 kV; excited-area size of 200e250 mm, measuring time of 2 min); the resultswere expressed as weight percentages and displayed as oxides.

According to the EDS data (Fig. 2e), out of the 63 prepareddiatom samples, 9 ranged between 97.2 and 99.6% SiO2, and be-tween 0.1 (detection limit) and 0.7% Al2O3; and 5 contained 95e97%SiO2, and about 1.5% Al2O3. These 14 specimens were analyzed foroxygen isotopes without additional cleaning. Among the remaining49 samples, 10 samples, mainly those extracted from the core in-terval dated to ~25.0e22.0 ka BP, were not suitable for isotopicanalysis due to small residual masses and high contamination(<90% SiO2 and up to 12% Al2O3). 22 other samples small in weightand containing 85e94% SiO2 and up to 3e5% Al2O3 were combinedinto seven samples, and then subjected to an additional HLS-cleaning together with the remaining 17 samples (91e95% SiO2and 1.9e4.3% Al2O3). After the HLS-cleaning procedure the Al2O3content decreased to 1.3e4.0% and the SiO2 content increased up to90e97%. SPT-cleaning only partly allowed a satisfactory separationbetween the terrigenous and diatom fractions. Finally, 38 sampleswere further processed and analyzed for their oxygen isotopecomposition.

3.5. Oxygen isotope analysis

The oxygen isotope composition of clean diatom material wasmeasured using the method published in Chapligin et al. (2010)with a PDZ Europa 2020 mass spectrometer (MS-2020; now sup-plied by Sercon Ltd., UK) at the AWI in Potsdam. Prior to isotopeanalysis, the samples were heated up to 1100 �C and cooled down to400 �C in approximately 7 h under a flow of He gas (inert Gas FlowDehydration (iGFD)) to remove any exchangeable groups in theamorphous biogenic structure containing oxygen (Chapligin et al.,2010). The dehydrated samples (about 2 mg) were then fullyreacted by laser fluorination under BrF5 atmosphere to liberate O2(Clayton and Mayeda, 1963), which was then separated from by-products and directly measured against an oxygen reference sam-ple of known isotopic composition. The working standard BFC(Chapligin et al., 2011) was used (this study: d18O¼þ28.8 ± 0.27‰;n ¼ 15; inter-laboratory comparison: þ29.0 ± 0.3‰). The oxygenisotope composition of diatom silica is expressed on the delta scalein per mil (‰). The final d18O value of the diatom sample is calcu-lated relative to V-SMOW standard with a long-term analyticalreproducibility of ±0.25‰ (Chapligin et al., 2010).

3.6. Contamination assessment and correction of isotopemeasurements

Due to potential changes in d18O values caused by anycontamination left in purified samples, the measured d18O signals

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 6: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e126

were corrected using a geochemical mass-balance approach(Brewer et al., 2008; Swann and Leng, 2009; Chapligin et al., 2012):

d18Ocorrected ¼ (d18Omeasured � d18Ocontamination $ ccontamination/100)/(cdiatom/100), (2)

where d18Omeasured is the original measured value d18O of theanalyzed sample, d18Ocorrected is the measured d18O value correctedfor contamination, d18Ocontamination ¼ þ10.5‰ and represents theaverage d18O value of the analyzed terrigenous samples (heavyfractions from all core samples merged into three samples(540e720 cm, 695e830 cm, 840e895 cm) to gain enough materialfor both EDS and isotope analyses) and ccontamination and cdiatom arethe percentages of contamination and diatommaterial respectivelywithin the analyzed sample. The percentage of contamination iscalculated by the EDS measured Al2O3 content of the individualsample divided by the average Al2O3 content of the contamination(14.2% in heavy fractions). The percentage of diatom material iscalculated as (100% e ccontamination).

Seven samples with more than 2.5% Al2O3 according to EDSanalysis (Fig. 2e)werediscarded fromfurther interpretationbecauseabove this limit the geochemicalmass-balance correction techniquecould cause excessive shifts in d18O values (Chapligin et al., 2012).

4. Results

4.1. Water isotope analysis

Water isotope data for the Baikal region includes the results ofstable isotope analysis of water samples collected in summer 2011published by Kostrova et al. (2013a) and new data collected duringthemonitoring period between 2011 and 2013. All results of isotopeanalyses are summarized in Table 1 and presented in a d18OedDdiagram (Fig. 3) with respect to the Global Meteoric Water Line(GMWL; dD ¼ 8 $ d18O þ 10), in which fresh surface waters (Craig,

Fig. 3. d18OedD diagram for Lake Kotokel and rivers connected to the lake as well asrain and snow water (sampled in Irkutsk) collected between 2011 and 2013 (compareTable 1). Additionally, GNIP data for Irkutsk precipitation, the Global Meteoric WaterLine as well as an evaporation line for Lake Kotokel water are given.

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

1961) and precipitation (Rozanski et al., 1993) are correlated on aglobal scale. The recent stable isotope composition of the surfaceLake Kotokel water during sampling periods variedbetween �10.8‰ and �13.7‰ for d18O and from �101.2‰to �114.7‰ for dD. The d excess changes from �4.7‰ to �14.8‰with a mean value of �9.1‰. All rivers draining into Lake Kotokelshow a mean isotopic composition of around �19.7‰ in d18O,and �144.5‰ in dD (mean d excess around þ12.7‰). Baikal regionrain samples are characterized by mean values of �11.8‰ for d18O,and �95.0‰ for dD (the d excess is þ0.1‰) whereas snow samplesdisplay mean d18O, dD and d excess values of �27.7‰, �210.5‰and þ11.1‰, respectively.

4.2. Influence of TMSR on the oxygen isotope composition

Fossil (PL) and recent (B1) diatom samples were initiallyanalyzed by EDS. The PL-sample contained 98.8 ± 0.4% (n¼ 7) SiO2,and 0.1 ± 0.1% (n ¼ 7) Al2O3. Concentrations of SiO2 and Al2O3 were99.4 ± 0.3% (n ¼ 7), and 0.01 ± 0.03% (n¼ 7), respectively in the B1-sample. Both samples were successively processed by using SPT forHLS (Morley et al., 2004; Chapligin et al., 2012) and TMCS for TMSR(Kalmychkov et al., 2005). The oxygen isotope composition for thePL-sample was þ22.2‰ before the questioned purificationmeasures were taken and þ22.4‰ versus þ22.0‰ after SPT- andTMCS-treatments, respectively. The d18O values for the B1-samplewereþ21.5‰ before andþ21.9‰ after HLS andþ21.5‰ after TMSR.

4.3. Diatom oxygen isotope analysis

The d18Ocorrected values from the investigated segments(842e892 and 500e715 cm depth) within the KTK2 core varybetween þ26.7 and þ31.2‰ (Fig. 2d). In general, the d18Ocorrectedvalues for the diatom samples follow the same trend as thed18Omeasured values, but are on average about 1.4‰ higher for thepart of the core spanning the time period from ~24.6 to 22.9 ka BP(7.4% average contamination) and approximately 2.0‰ higher forthe interval between ~16.7 and 12.7 ka BP (10.2% averagecontamination). In the following, we use the corrected data ofdiatom samples containing less than 2.5% Al2O3 (or >15% contam-ination) for interpretation.

Obtained d18Odiatom values (Fig. 2d) from the LGM and lateglacial intervals generally match the range of other lacustrinediatom records (þ15 to þ40‰) (Leng and Barker, 2006 and refer-ences therein; Swann et al., 2010; Mackay et al., 2011; Hernandezet al., 2013) as well as that from Holocene Lake Kotokel diatoms(Kostrova et al., 2013a, b). The interval between ~24.6 and23.0 ka BP displays relatively constant d18Odiatom values aboutof þ30.1‰ with a visible spike of þ27.8‰ at ~23.2 ka BP. Themaximal d18Odiatom value (þ31.2‰) is registered at ~14.4 ka BP.After a sharp drop to þ26.7‰ at ~14.2 ka BP e the absolute mini-mum in the d18O record e and a subsequent rise to þ30.3‰, thed18O values gradually increase toþ31.0‰ from ~14.1 to 13.7 ka BP. Adecline of 2.5‰ in the KTK2 isotope record is observed at~13.7e13.5 ka BP. After rapid increase to þ31.0% at 13.4 ka BP asmaller minimum with a d18O value of þ29.8‰ occurs at~13.3 ka BP. The interval ~13.2e12.5 ka BP is characterized byrelatively stable d18O values ranging between þ30.4 and þ30.9‰.At ~12.3 ka BP there is another minimum in d18O with þ29.0‰.After ~12.3 ka BP, d18O values increase to þ31.1‰ in the upper partof the core. In general, a gradual enrichment of ~3.5‰ in d18O isobserved in diatom frustules during the period ~17.0e11.5 ka BP(Fig. 2d). Here, the linear correlation between d18Odiatom and cali-brated ages (in yr) yields: d18Odiatom ¼ �0.0005 yr þ36.37‰(R2 ¼ 0.23).

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 7: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e12 7

5. Discussion

5.1. The effect of TMSR on the oxygen isotope signal

A variety of techniques have been developed for the extractionand cleaning of diatoms (see Chapligin et al., 2012 for more details)for stable isotope analysis. The purification procedure used in thisstudy involves a complex cleaning protocol which was firstreported by Kalmychkov et al. (2005, 2007) and recently applied byKostrova et al. (2013a, b).

Any chemical reaction including the applied acid exposure orTMSR may probably cause a change on the frustules' surface asrupture of siloxane (≡SieOeSi≡) bonds and therefore does notguarantee a good preservation of the oxygen-isotope signal duringpreparation of a sample for isotope analysis. Brandriss et al. (1998)noted the effect of hot acids on fresh diatoms causing the increaseof d18O values of 1.8 and 5.3‰, whereas the same treatment had noeffect on fossil diatoms.

The PL-sample shows a ±0.2‰ difference in d18O values beforeand after SPT- and TMCS-treatment was undertaken. For sample B1no difference in the oxygen isotope composition was observedbefore and after TMSR, whereas after carrying out HLS a slightdifference of þ0.4‰ was found. For both samples the differencebetween d18O value before and after SPT- and TMCS-processingwaswithin the instrument's margin of error (SD ¼ ±0.25‰). Hence,these changes are not significant and do not relate to potentialstructural changes in diatoms. Thus, TMSR does not provoke amodification of the oxygen isotope composition and can be appliedfor the extraction and cleaning of diatom from terrigenous matteras an alternative to HLS.

5.2. Potential controls for the diatom isotope record

Variations in d18Odiatom values of lacustrine sediment are mainlycontrolled by changes in water temperature and/or the oxygenisotope composition of the corresponding lake water (d18Olake)which is affected by alterations in lake hydrology and d18O values ofatmospheric precipitation (d18Op; Leng and Marshall, 2004; Lengand Barker, 2006).

5.2.1. Water temperatureDuring the LGM and late glacial Lake Kotokel summer water

temperatures were certainly colder than the modern ones, whichyield an average of 18 �C from May to October (Kostrova et al.,2013a and references therein). Assuming a LGM/late glacialaverage summer water temperature around 10e12 �C, this wouldincrease d18Odiatom values by ~1.6e1.2‰ due the temperature-dependency of isotope fractionation (�0.2‰/�C; Swann andLeng, 2009). Since the mean d18Odiatom value for the LGM/lateglacial period is about 1.8‰ higher than for the Holocene(Kostrova et al., 2013a), the average offset between the LGM/lateglacial and the Holocene oxygen isotope composition could thusbe mostly explained by water temperature changes. Diatom as-semblages from the LGM and late glacial intervals of the KTK2 core(Fig. 2b; Bezrukova et al., 2010) mainly involve species that usuallyprevail during the summer period in a wide range of water tem-peratures, i.e. between ~7e8 �C and 18e20 �C (Kuz'mich, 1988;Smol, 1988; Popovskaya et al., 2002; Barinova et al., 2006;Finkelstein and Gajewski, 2008; Rühland et al., 2008; Kuzminet al., 2009; Wang et al., 2012b). Assuming that water tempera-ture is the primary control for d18Odiatom values from the LGM/lateglacial Lake Kotokel record, then the high-frequency variability onthe order of 4.5‰ would result in a theoretical variation of 22.5 �Cin summer water temperatures. Because the modern water tem-perature range in Lake Kotokel is only about 14 �C (Kuz'mich,

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

1988), a change of 22.5 �C during the shorter summer period inthe colder LGM and late glacial is likely unrealistic. Consequently,water temperature variations can only partly explain the change ofd18Odiatom values during the LGM and late glacial. Hence, otherfactors, i.e. changes in the oxygen isotopic composition of the lakewater, need to be considered.

5.2.2. Hydrology and meteorological backgroundThe interpretation of diatom isotope records from lake sedi-

ments requires an understanding of the modern hydrologicalconditions and should take possible past changes in the hydro-logical regime into account.

Recent works (Kostrova et al., 2013a, b) have demonstrated thatalterations in the oxygen isotope composition of the Lake Kotokelwater throughout the Holocene were mainly caused by (1) changesin d18Op values in the Kotokel catchment, as well as precipitationamounts and origin; (2) supply of snow melt water to the lake and(3) varying evaporative effects. The following paragraphs aim todiscuss the effects of isotope hydrology and local meteorology onthe Lake Kotokel d18Odiatom record.

Stable isotope data for precipitation collected in Irkutsk duringthe time period from 2011 to 2013 show distinct seasonal varia-tions. Summer air masses in the Baikal region display relativelyhigh d18Op and dDp values of up to�4.7‰ and�47.1‰, respectively,while winter air masses generally display a lower isotopiccomposition: in December and January d18Op and dDp values maydrop to �39.5‰ and �318.5‰, respectively (Table 1). The linearcorrelation between the isotopic composition of precipitation andmean daily air temperatures with precipitation events derived fromthe global summary of the day data sets provided by the NationalOceanic and Atmospheric Administration (NOAA) and the NationalClimatic Data Center (NCDC; available at: www.ncdc.noaa.gov/data-access/quick-links) yields: d18Op ¼ 0.57$Tair � 19.33‰(R2 ¼ 0.74) for oxygen and dDp¼ 4.16$Tair� 149.52‰ (R2¼ 0.71) forhydrogen, reflecting the seasonal changes of precipitation isotopedata. Seal and Shanks (1998) calculated a positive relationshipof þ0.36‰/�C for monthly atmospheric air temperature and d18Opfor the Lake Baikal region. Since presently Tair has a stronger effecton d18Olake as compared to Tlake (Kostrova et al., 2013a), we need toconsider that d18Odiatom might be related to air temperatures ratherthan to lake temperatures also during the LGM and late glacialintervals.

Three-year observations show that the isotope composition ofthe surface Lake Kotokel water has not undergone significantchanges and ranges in a relatively narrow span around averagevalues of�12.2‰ for d18O,�107.2‰ for dD and is characterized by alowmean d excess of�9.1‰ (Table 1). These values indicate a well-mixed lake water environment, which is encompassed by a lack ofsignificant temperature differences between surface and bottomwater due to constant wind mixing of the water mass (Kuz'mich,1988). Lake Kotokel water isotope data plot to the right andbelow of the GMWL (Fig. 3), following the linear dependencedD ¼ 5.0$d18O e 46.3 (R2 ¼ 0.92), a so called Evaporation Line (EL),indicating that the lake water is influenced by evaporative enrich-ment. The intersection point of EL and GMWL is at �18.5‰ for d18Oand at �140‰ for dD and, hence, quite similar to but a little higherthan the mean isotopic composition of the inflow rivers (i.e.around �20‰ in d18O, and �145‰ in dD, mean d excessaround þ13‰; Table 1), suggesting that Lake Kotokel is predomi-nantly fed by snow melt water from the hinterland and thus byprecipitation from higher altitudes characterized by lighter d18Oand dD values (Kostrova et al., 2013a, b).

As shown above, the River Istok connected to Lake Kotokel(Fig. 1b) can act either as outflow or as inflow depending onwhether or not the River Kotochik is filled by melt water. We

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 8: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e128

calculated the amount of Lake Kotokel water in the River Kotochikbelow the confluence of the River Istok using geochemical mass-balancing for two-component mixing systems:

d18OKA ¼ cKTK $ d18OKTK þ cKB $ d18OKB, (3)

Fig. 4. A and B: SEM pictures of cleaned diatoms from the late glacial part of LakeKotokel sediments. Traces of dissolution are absent. Fragments of diatoms are theresult of mechanical effects during sample preparation.

where d18OKA is the d18O value for the River Kotochik below theconfluence with the River Istok, d18OKTK is the d18O value for LakeKotokel at the time of sampling, d18OKB relates to the River Kotochikabove the confluence of the River Istok, cKTK is the share of the LakeKotokel water and cKB ¼ (1e cKTK) is the share of the River Kotochikwater at the confluence of the River Istok.

The d18O and dD values for the River Kotochik above theconfluence of the River Istok are �22.0‰ and �160.7‰, respec-tively. Water from the River Istok at the end of July 2011 shows arelatively constant isotopic composition of d18O around�20.4‰, dDaround �151.0‰ and d excess around þ12.3‰ (Table 1). At thesame time, the mean d18O and dD values of the River Kotochikwater below the confluence with the River Istok are �21.6‰and �157.5‰, respectively. This means that, according to ourcalculation about 6% of the River Istok water originates from LakeKotokel, whereas the major part of its water comes from the RiverKotochik. This indicates that the water from the River Kotochikmight flow into the lake through the Istok. The isotopic data ob-tained for the Istok-Kotochik river system in 2012 and 2013 (Fig. 1b;Table 1) show a completely different picture. The stable isotopecomposition of the River Istok water (mean values for d18Oaround �12.2‰; dD around �108‰ and d excess around �10.9‰)is similar to that of Lake Kotokel water (Table 1) pointing to LakeKotokel water as the origin of the River Istok. In accordance withthemixing calculation, about 80% of the Lake Kotokel water reachesthe confluence of the rivers Kotochik and Istok. At a distance ofabout 2 km downstream the Kotochik River, the lake water share isreduced to 35%. The results confirm that Kotokel is partly a shallowthrough-flow lake and partly acts as a closed system while under-going significant evaporation.

5.2.3. Species-effect and dissolutionAdditionally, species (or vital) and dissolution effects have to be

considered when interpreting diatom isotope data (Leng andBarker, 2006; Mackey et al., 2013). Only one study identifiedsmall species effects on marine d18Odiatom (Swann et al., 2008),whereas others found no significant changes of the isotope signal(Moschen et al., 2005; Swann et al., 2006; Chapligin et al., 2012). Asthis has not been fully investigated in lake environments the pos-sibility cannot be completely excluded. The applied TMSR-cleaningprocedure has no potential to separate samples into individualspecies for isotope analyses and thus to take into account specific-species changes. The diatom assemblages of the KTK2 core duringthe LGM and late glacial show some variations. Throughout theinterval between ~24.7 and 21.9 ka BP, S. pinnata agg. valvesdominate (up to 82%; Fig. 2b). Biogenic silica from other species inthis period is relatively low and, thus, possible effects of changingassemblage composition on d18Odiatom values are also assumed aslow. The presence of varying amounts of major species as plank-tonic A. granulata, E. arenaria var. arenaria and benthic P. brevistriata(Fig. 2b; Bezrukova et al., 2010) in the period ~17.0e12.7 ka BP maycause some changes in the isotopic signature possibly in conse-quence of different species-specific fractionation patterns duringdiatom growth (Chapligin et al., 2012). However, any effect relatedto blooming in a different season or living in different water depthscould be excluded becausemost Lake Kotokel diatoms predominatein summer time when there are no significant differences in the

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

water isotope composition (Table 1). Additionally, despite a con-stant trend in the diatom assemblage (from a dominatedP. brevistriata to A. granulata) within this time period no overalltrend is visible in the diatom isotope record.

An increase of d18Odiatom values takes place if dissolution of di-atoms occurs at pH 9.0 (Moschen et al., 2006). As the modern pHmeasured at Lake Kotokel varies between 6.8 and 7.3 (Kuz'mich,1988) and the pH reconstructed on the basis of diatom analysis ofthe KTK2 core ranges from 4.5 to 8.0 (Bezrukova et al., 2010), it isunlikely that dissolution had an effect on the obtained d18Odiatomvalues. Additionally, due to the shallowness of Lake Kotokel and thegood preservation of Kotokel diatoms under the SEM (Fig. 4),dissolution effects are much less significant than for the deeperLake Baikal (Ryves et al., 2003; Battarbee et al., 2005). Conse-quently, it is unlikely that any species-effect or dissolution processchanged the isotope signal. Based on the argumentation of thepreceding chapters, we believe that the dominant control ond18Odiatom in a colder and drier LGM/late glacial environment, ascompared to the Holocene, is evaporation. However, the balancebetween lake and air temperature changes will certainly affectlong-term trends, whereas short-term events to lower d18Odiatom

might be related to an increased water supply to the lake, i.e. due tochanging precipitation rates or melt water pulses from higheraltitudes.

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 9: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e12 9

5.3. The LGM and late glacial diatom isotope record and lacustrineenvironment

According to palaeoclimate studies based on high-resolutionproxy records from Lake Kotokel (Bezrukova et al., 2008, 2010,2011; Tarasov et al., 2009; Müller et al., 2013) and surroundingpeat bogs (Shichi et al., 2009), the LGM/late glacial climate andenvironment around the lake and its ecosystem underwent sig-nificant changes, mainly caused by the long- and short-term vari-ations of temperature and atmospheric precipitation. Pollen-basedreconstructions demonstrate a generally colder and drier thanpresent climate over the LGM and late glacial interval (Bezrukovaet al., 2010; Müller et al., 2013).

As mentioned above, conditions unfavourable for diatomdevelopment occurred between about 31.9 and 24.7 ka BP and~22.0e17.0 ka BP (Fig. 2b and c; Bezrukova et al., 2010) whileostracod shell fragments were discovered in Lake Kotokel withinthe time interval ~26.8e19.1 ka BP (Müller et al., 2013). The diatomisotope record from Lake Kotokel between about 24.7 and

Fig. 5. Comparative plots for periods 17e11.5 and 25e23 ka BP showing (a) correctedd18O isotopic records from Lake Kotokel (this study); (b) the pollen-inferred scores ofdominant biomes (Bezrukova et al., 2010) as palaeoclimatic indicators in the LakeBaikal region; along with (c) the NGRIP d18O record from Greenland ice (Svenssonet al., 2008) as an indicator of the Northern Hemisphere (NH) air temperature and(d) the d18O records from Chinese stalagmites D4 and MSD (Wang et al., 2001; Yuanet al., 2004) as an indicator of the Pacific monsoon intensity; (e) the NH summerinsolation at 60� N (Berger and Loutre, 1991). Grey bars indicate the approximateposition of the Younger Dryas (YD), Allerød (AL), Bølling (BO) and Meiendorf (MD)identified by pollen and diatom records from Lake Kotokel (Bezrukova et al., 2010).

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

23.0 ka BP is characterized by a relatively high oxygen isotopecomposition of þ30.1‰ (Fig. 5a) interrupted by an abrupt eventof þ28.7‰ at ~23.2 ka BP (Fig. 5a) which is also detected in theNGRIP record (Fig. 5d; Svensson et al., 2008). Over this period, thediatom complex consists of small benthic S. pinnata agg., Staurosiraconstruens agg., P. brevistriata and O. martyi with a nearly completedominance of the S. pinnata agg. (Fig. 2b; Bezrukova et al., 2010)suggesting a short open-water season, a turbid environment and ahigh level of erosion (Smol, 1988). At the same time, the increase incoarse-grained sand particles together with high percentages ofRanunculaceae pollen grains in the Kotokel sediments also pointsto intensified soil erosion and a much shorter than present distancebetween the coring site and the shoreline, and thus a reduced lakearea (Müller et al., 2013). Taking into account the substantiallydrier-than-present LGM climate of the Baikal region (Goldberget al., 2010) and lower-than-present mean July air temperaturesof þ13 �C (Müller et al., 2013), relatively high d18Odiatom values inthe interval between ~24.7 and 23.0 ka BP (Fig. 5a) were mostlikely caused by enhanced evaporation playing a major role incontrolling changes in lake water d18O during this period. Along thenorth-eastern coast the underwater river channel fills are indica-tive of a lower lake level prior to 15 ka BP (Zhang et al., 2013),confirming our assumption. At the same time, enhanced evapora-tion linked to increased aridity in inner Asia (Karabanov et al.,2004; Goldberg et al., 2010) resulted in a lowering of the waterlevel of lakes in Mongolia (Prokopenko et al., 2005) and China(Kramer et al., 2010).

Changes in the lithology and diatom records from Lake Kotokelafter ~17 ka BP (Fig. 2a and b) indicate the late-glacial climateamelioration in the Baikal region. According to pollen data, the firstmajor warm episode after the LGM occurred at ~14.7 ka BP asindicated by a marked expansion of trees and shrubs and areduction of steppe vegetation (Fig. 5b; Tarasov et al., 2009;Bezrukova et al., 2010). In the Lake Kotokel sediments diatoms S.pinnata and P. brevistriata preferring relatively cold water(Finkelstein and Gajewski, 2008) are replaced by comparativelywarm-water species A. granulata and A. ambigua (Fig. 2b;Popovskaya et al., 2002; Bezrukova et al., 2010). Our d18Odiatom re-sults (Fig. 5a) point to an increase of about 2‰ between 16.7 and14.4 ka BP, which could be related to the post LGM warming trend.This is in line with an increase in sea surface temperatures in themid-latitudes of the western North Atlantic to 15e17 �C at~15.0 ka BP (Rodrigues et al., 2010), the strengthening of the EastAsian summer monsoon around 14.8 ka BP (Wang et al., 2001,2012a; Yuan et al., 2004) and an increasing precipitation amountreconstructed in the Baikal region (Demske et al., 2005; Tarasovet al., 2009).

Between ~15 and 11.5 ka BP, the isotope composition of LakeKotokel diatoms (Fig. 5a) remains at a high level reaching d18Odiatomvalues of up to þ31.2‰, interrupted by several single spikes oflower isotope composition (i.e. þ26.7‰ at ~14.3 ka BP; þ28.5‰ at~13.5 ka BP; þ29.0‰ at ~12.3 ka BP). The overall high level ofd18Odiatom values in a rather cold late glacial interval seems not tocorrespond to air temperature changes since (1) low Tair would leadto low d18Olake and thus low d18Odiatom, (2) several short-termwarmand cool phases known from the pollen-based temperature re-constructions (Tarasov et al., 2009; Bezrukova et al., 2010) andNGRIP record (Fig. 5c; Svensson et al., 2008) are not clearly re-flected in the d18Odiatom record. The most appropriate way toexplain the high d18Odiatom values is a highly evaporative LakeKotokel system with little precipitation reaching this highly conti-nental site, even in phases of climate amelioration and increasingprecipitation amounts (Tarasov et al., 2009; Goldberg et al., 2010).Most likely, while receiving less precipitation, Lake Kotokel acted asa closed-system basin during longer periods per year.

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 10: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e1210

A general warming corresponding to the MD interstadialobserved in the pollen data (Litt and Stebich, 1999; Stebich et al.,2009; Bezrukova et al., 2010; Sirocko, 2012) and in the NGRIP re-cord (Fig. 5c; Svensson et al., 2008) during ~14.5e14.0 ka BP mighthave caused themelting of perennial snowfields or isolated ice capsin the hinterland of Lake Kotokel which could be a source of waterwith low d18O. This melt water input consequently explains thespike around ~14.3 ka BP in the Lake Kotokel diatom isotope record(Fig. 5a). Shichi et al. (2013) also note a remarkable expansion ofriver floodplain areas near Lake Baikal and the release of water tothe lake basin in response to melting of mountain glaciers andfrozen soils under warmer conditions since ~15 ka BP. A progressiveincrease in the abundance of planktonic A. granulata, A. ambigua, C.ocellata diatoms in the Lake Kotokel sediments and a noticeabledecrease in benthic species (Fig. 2b) indicate a longer open-waterseason and a deepening of the lake during ~14.5e14.0 ka BP(Bezrukova et al., 2010). Around ~14.3 ka BP the total diatom con-centration reaches its maximum (Fig. 2c). At the same time,increased amounts of precipitation in north-east China and thaw-ing of local permafrost caused a rise in lake levels (Stebich et al.,2009).

The diatom oxygen isotope record from Lake Kotokel shows asecond abrupt event (d18Odiatom ¼ þ28.5‰) registered between BOand AL interstadials at ~13.5 ka BP (Fig. 5a) that could be corre-lated with the OD cooling (Yu and Eicher, 2001; Stebich et al.,2009). There is some evidence for this event in the KTK2 pollenand diatom records (Bezrukova et al., 2010) and a relatively lowvalue of þ24.9‰ was also detected in the d18O diatom record fromLake Baikal (Mackay et al., 2011) at ~13.4 ka BP. These declines inboth records are in line with a short-term negative oxygen isotopeshift in the NGRIP ice core (Fig. 5c; Svensson et al., 2008). A cli-matic deterioration around 13.6e13.4 ka BP has been clearlydocumented in several high-resolution records from Europe,North America and Asia (Yu and Eicher, 2001; Lauterbach et al.,2011; Zhao et al., 2013), related to a shift in atmospheric circula-tion patterns and instantaneous environmental responses to thischange. At this time, the replacement of spruce and fir forests byshrubby alder and dwarf birch associations around Lake Kotokelsuggests cooler and moderately wet conditions with300e320 mm/y precipitation likely prevailed in winter (Tarasovet al., 2009; Bezrukova et al., 2011). The alder communitycommonly grows within wet, but swamp-devoid areas in wintercovered by a thick snow layer (Gerasimov, 1966).

At ~13.5e12.8 ka BP the Baikal region climate becamewarm andwet according to pollen-based quantitative reconstructions(Tarasov et al., 2007, 2009). In Lake Kotokel the sedimentation typechanged frommainlyminerogenic to organogenic (Bezrukova et al.,2008). The almost complete dominance of planktonic A. granulataand A. ambigua diatom species and benthic Staurosira construensagg. in the sediments (Fig. 2b) indicates the deepening of the lake(Bezrukova et al., 2010). However, the d18Odiatom values are at a highlevel aroundþ30.5‰ until ~12.8e12.7 ka BP showing no significantresponse to thementioned environmental changes. At that time theamount of precipitation reached ~400 mm, comparable to presentlevels (Tarasov et al., 2009), and the stalagmite oxygen isotope re-cord from China demonstrates a strengthening of the East Asiansummer monsoon between ~13.5 and 13 ka BP (Fig. 5d; Yuan et al.,2004; Wang et al., 2012a).

The YD stadial had global effects on the environment includingthe Baikal regionwhere it has been clearly identified between ~12.7and 11.65 ka BP (Bezrukova et al., 2010). A decrease in mass accu-mulation rates of total organic carbon and in concentrations ofbiogenic silica and diatoms took place in Lake Baikal (Mackay,2007; Mackay et al., 2011 and references therein). Around LakeKotokel a reduction of forest vegetation and an expansion of steppe

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

and tundra associations occurred (Fig. 5b; Bezrukova et al., 2010).Pollen-based reconstructions suggest lower than present temper-atures (by 2e4 С� in July and by 8 С� in January) and lower pre-cipitation (by 100 mm/yr) in the region for this period (Tarasovet al., 2007, 2009). However, the Lake Kotokel d18Odiatom record(Fig. 5a) only shows a minor decline of about 1.3‰~12.8e12.4 ka BP, which is accompanied by the appearance of cold-water E. arenaria var. arenaria (Fig. 2b) and in line with lower airtemperatures in the region during the YD (Tarasov et al., 2009). Theoxygen isotope record from Lake Baikal demonstrates a similardistribution of d18Odiatom values between ~12.8 and 11.6 ka BP(Mackay et al., 2011). The YD d18Odiatom spike is also roughly syn-chronous with the oxygen isotope records from Greenland (Fig. 5c;Svensson et al., 2008) and from China (Fig. 5d; Yuan et al., 2004).However, neither a preceding warm AL nor a cold YD are readilyvisible in the Lake Kotokel d18Odiatom record.

The overall trend of the Lake Kotokel isotope record spanningthe period between ~15.0 and 11.5 ka BP (Fig. 5a) is similar to that ofJune insolation at 60� (Fig. 5e; Berger and Loutre, 1991) suggestingthat the diatom d18O value is a sensitive parameter responding tochanges in solar activity. Despite a visible reaction of the late glacialoxygen isotope record from Lake Kotokel to regional changes in airtemperatures as single spikes, the effect of Tair on d18O of the lakewater was smaller in comparison to evaporation. The meand18Odiatom value for the LGM/last glacial periods has been deter-mined asþ30.1‰which is about 1.8‰ higher than for the Holocene(Kostrova et al., 2013a) suggesting that during the LGM and lateglacial, Lake Kotokel was a strongly evaporative system. This is inline with a strengthening of the Siberian High likely causingincreased aridity in the catchment of Lake Kotokel and alteringfluvial input into the lake. Apparently, LGM/late glacial winterswere characterized by less snow than present, which could havelead to a reduction of river input fromhigher altitudes and hence anisotope enrichment of the lake water.

6. Conclusions

This paper presents the newly obtained oxygen isotope recordon fossil diatoms retrieved from a sediment core from Lake Kotokeland used for reconstructing climate and environment dynamics inthe Baikal region during the LGM and late glacial times. The presentstudy has shown the possibility of applying the TMSR as an alter-native technique for extracting and cleaning diatoms for oxygenisotope analysis.

Our study provides important information on environment andclimate history of the Baikal region in central Asia during the LGMand the late glacial intervals indicating that the temperature andmoisture variations influenced the oxygen isotope values in fossildiatoms from Lake Kotokel. The broad similarities between the LakeKotokel diatom isotope record and Northern Hemisphere solarinsolation show that d18Odiatom responded well to the insolation-induced temperature changes. The combination of the d18Odiatomrecord with pollen and diatom data allows discussion of regionalenvironmental changes and their responses in the lake ecosystemin more detail. We suggest that Lake Kotokel acted for most of thelate glacial as a strongly evaporative system. Enhanced evaporationand prevalent summer precipitation in the region are the mainreasons for overall high d18Odiatom values up toþ31.2‰ during a dryand cold LGM/late glacial. Enhanced melt water input from thehinterland (i.e. at ~14.3 ka BP) as a reaction to warming pulses wasdetected as minima in the d18O record of Lake Kotokel. This studydemonstrates the complex interplay of global and regional-scalehydrological and climatological factors controlling the d18Odiatom,especially in a cold glacial environment substantially different frompresent day conditions.

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 11: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e12 11

Acknowledgements

This study is a contribution to the research program “BridgingEurasia” supported by German Research Foundation (DFG; grantsTA-540/4, TA 540/5) and the Russian Foundation for Basic Research(RFBR; grant 12-05-00476a). The study was also partly supportedby the DFG grants Me-3266-3-1, Ме-3266-5-1 and by the GermanAcademic Exchange Service (DAAD; grant A-13-00095). The au-thors would like to thank Helga Kemnitz from GFZ for her SEMsupport. Additional thanks are owed to Gennady Kalmychkov andValery Bychincky from VIG SB RAS for their active participation inthe sampling of Lake Kotokel water and for discussions about theinfluence of TMSR on oxygen isotope data. Special thanks are alsodue to Lutz Sch€onicke (AWI), Yury Suslikov and Pavel Borovets(Arnika Industrial Service) for their technical support and to GilesShephard for polishing English. We sincerely thank three anony-mous reviewers for valuable and constructive comments and sug-gestions that helped improve an earlier version of this manuscript.

References

Barinova, S.S., Medvedeva, L.A., Anisimova, O.V., 2006. Diversity of Algal Indicatorsin the Environmental Assessment. Pilies Studio, Tel Aviv, p. 498.

Battarbee, R.W., Mackay, A.W., Jewson, D.H., Ryves, D.B., Sturm, M., 2005. Differ-ential dissolution of Lake Baikal diatoms: correction factors and implications forpalaeoclimatic reconstruction. Global and Planetary Change 46, 75e86.

Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 millionyears. Quaternary Science Reviews 10 (4), 297e317.

Bezrukova, E.V., Krivonogov, S.K., Takahara, H., Letunova, P.P., Shichi, K.,Abzaeva, A.A., Kulagina, N.V., Zabelina, Yu, S., 2008. Lake Kotokel as a stratotypefor the Late Glacial and Holocene in Southeastern Siberia. Doklady Earth Sci-ences 420 (4), 658e663.

Bezrukova, E.V., Tarasov, P.E., Solovieva, N., Krivonogov, S.K., Riedel, F., 2010. Lastglacial-interglacial vegetation and environmental dynamics in southern Siberia:chronology, forcing and feedbacks. Palaeogeography, Palaeoclimatology,Palaeoecology 296, 185e198.

Bezrukova, E.V., Tarasov, P.E., Kulagina, N.V., Abzaeva, A.A., Letunova, P.P.,Kostrova, S.S., 2011. Palynological study of Lake Kotokel’ bottom sediments.Russian Geology and Geophysics 52, 457e464.

Bo€es, X., Piotrowska, N., Fagel, N., 2005. High-resolution diatom/clay record in LakeBaikal from grey scale, and magnetic susceptibility over Holocene and Termi-nation I. Global and Planetary Change 46, 299e313.

Brandriss, M.E., O'Neil, J.R., Edlund, M.B., Stoermer, E.F., 1998. Oxygen isotopefractionation between diatomaceous silica and water. Geochimica et Cosmo-chimica Acta 62, 1119e1125.

Brewer, T.S., Leng, M.J., Mackay, A.W., Lamb, A.L., Tyler, J.J., Marsh, N.G., 2008.Unravelling contamination signals in biogenic silica oxygen isotope composi-tion: the role of major and trace element geochemistry. Journal of QuaternaryScience 23 (4), 321e330.

Chapligin, B., Meyer, H., Friedrichsen, H., Marent, A., Sohns, E., Hubberten, H.-W.,2010. A high-performance, safer and semi-automated approach for the d18Oanalysis of diatom silica and new methods for removing exchangeable oxygen.Rapid Communications in Mass Spectrometry 24, 2655e2664.

Chapligin, B., Leng, M.J., Webb, E., Alexandre, A., Dodd, J.P., Ijiri, A., Lücke, A.,Shemesh, A., Abelmann, A., Herzschuh, H., Longstaffe, F.J., Meyer, H.,Moschen, R., Okazaki, Y., Rees, N.H., Sharp, Z.D., Sloane, H.J., Sonzongi, C.,Swann, J.E.A., Sylvestre, F., Tyler, J.J., Yam, R., 2011. Inter-laboratory comparisonof oxygen isotope compositions from biogenic silica. Geochimica et Cosmo-chimica Acta 75, 7242e7256.

Chapligin, B., Meyer, H., Bryan, A., Snyder, J., Kemnitz, H., 2012. Assessment of pu-rification and contamination correction methods for analysing the oxygenisotope composition from biogenic silica. Chemical Geology 300e301, 185e199.

Clayton, R., Mayeda, T., 1963. The use of bromine pentafluoride in the extraction ofoxygen from oxides and silicates for isotopic analysis. Geochimica et Cosmo-chimica Acta 27, 43e52.

Cook, C.C., Leng, M.J., Jones, R.T., Langdon, P.G., Zhang, E., 2012. Lake ecosystemdynamics and links to climate change inferred from a stable isotope and organicpalaeorecord from a mountain lake in southwestern China (ca. 22.6e10.5 cal kaBP). Quaternary Research 77, 132e137.

Craig, H., 1961. Isotopic variations in meteoric waters. Science 133, 1702e1703.Demske, D., Heumann, G., Granoszewski, W., Nita, M., Mamakowa, K., Tarasov, P.E.,

Oberhansli, H., 2005. Late glacial and Holocene vegetation and regional climatevariability evidenced in high-resolution pollen records from Lake Baikal. Globaland Planetary Change 46, 255e279.

Diefendorf, A.F., Patterson, W.P., Mullins, H.T., Tibert, N., Martini, A., 2006. Evidencefor high-frequency late Glacial to mid-Holocene (16,800 to 5000 cal yr B.P.)climate variability from oxygen isotope values of Lough Inchiquin, Ireland.Quaternary Research 65, 78e86.

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

Finkelstein, S.A., Gajewski, K., 2008. Responses of Fragilarioid-dominated diatomassemblages in a small Arctic lake to Holocene climatic changes, Russell Island,Nunavut. Canada. Journal of Paleolimnology 40, 1079e1095.

Florensov, N.A., 1960. Mezozoiskie i kainozoiskie vpadini Pribaikal'ya. Izdatel'stvoAkademii Nauk SSSR, Moscow-Leningrad, p. 258 (in Russian).

Galaziy, G.I. (Ed.), 1993. Baikal Atlas. Federal Agency for Geodesy and Cartography ofRussia, Moscow (in Russian).

Gerasimov, I.P. (Ed.), 1966. Natural Conditions and Natural Resources of the USSR:North of the European Part of the USSR. Nauka, Moscow, p. 452 (in Russian).

Goldberg, E.L., Chebykin, E.P., Zhuchenko, N.A., Vorobyeva, S.S., Stepanova, O.G.,Khlystov, O.M., Ivanov, E.V., Weinberg, E., Gvozdkov, A.N., 2010. Uranium iso-topes as proxies of the environmental history of the Lake Baikal watershed (EastSiberia) during the past 150 ka. Palaeogeography, Palaeoclimatology, Palae-oecology 294, 16e29.

Hernandez, A., Bao, R., Giralt, S., Barker, P.A., Leng, M.J., Sloane, H.J., S�aez, A., 2011.Biogeochemical processes controlling oxygen and carbon isotopes of diatomsilica in Late Glacial to Holocene lacustrine rhythmites. Palaeogeography,Palaeoclimatology, Palaeoecology 299, 413e425.

Hernandez, A., Bao, R., Giralt, S., S�aez, A., Leng, M.J., Barker, P.A., Kendrick, C.P.,Sloane, H.J., 2013. Climate, catchment runoff and limnological drivers of carbonand oxygen isotope composition of diatom frustules from the central AndeanAltiplano during the Lateglacial and Early Holocene. Quaternary Science Re-views 66, 64e73.

Iler, R.K., 1979. The Chemistry of Silica: Solubility, Polymerization, Colloid andSurface Properties and Biochemistry of Silica. Wiley-Interscience Publication,New-York, Chichester, Bribano, Toronto, p. 866.

Karabanov, E., Williams, D., Kuzmin, M., Sideleva, V., Khursevich, G., Prokopenko, A.,Solotchina, E., Tkachenko, L., Fedenya, S., Kerber, E., 2004. Ecological collapse ofLake Baikal and Lake Hovsgol ecosystems during the Last Glacial and conse-quences for aquatic species diversity. Palaeogeography, Palaeoclimatology,Palaeoecology 209, 227e243.

Kashutina, M.V., Ioffe, S.L., Tartakovskii, V.A., 1975. Silylation of organic compounds.Russian Chemical Reviews 44 (9), 733e747.

Kalmychkov, G.V., Kostrova, S.S., Geletii, V.F., Tkachenko, L.L., Rakhlin, V.I., 2005.Method of separation of diatom frustules from bottom sediments for oxygenisotopic analysis and paleoclimatic reconstruction. Geochemistry International43, 1252e1254.

Kalmychkov, G.V., Kuz'min, M.I., Pokrovskii, B.G., Kostrova, S.S., 2007. Oxygen iso-topic composition in diatom algae frustules from Lake Baikal sediments: annualmean temperature variations during the last 40 Ka. Doklady Earth Sciences 413(2), 206e209.

Kostrova, S.S., Meyer, H., Chapligin, B., Kossler, A., Bezrukova, E.V., Tarasov, P.E.,2013a. Holocene oxygen isotope record of diatoms from Lake Kotokel (southernSiberia, Russia) and its palaeoclimatic implications. Quaternary International290e291, 21e34.

Kostrova, S.S., Meyer, H., Chapligin, B., Bezrukova, E.V., Tarasov, P.E., Kuz’min, M.I.,2013b. Reconstruction of the Holocene climate of Transbaikalia: evidence fromthe oxygen isotope analysis of fossil diatoms from Lake Kotokel. Doklady EarthSciences 451 (1), 732e736.

Kramer, A., Herzschuh, U., Mischke, S., Zhang, C., 2010. Late glacial vegetation andclimate oscillations on the southeastern Tibetan Plateau inferred from the LakeNaleng pollen profile. Quaternary Research 73, 324e335.

Kurita, N., Yoshida, N., Inoue, G., Chayanova, E.A., 2004. Modern isotope climatologyof Russia: a first assessment. Journal of Geophysical Research 109 (D03102),1e15.

Kuz'mich, V.N. (Ed.), 1988. Bioproduktivnost‘ evtrophnich ozer Irkana i Kotokelbasseina ozera Baikal. Sbornik Nauchnih Trudov GosNIOPH, vol. 279, p. 150 (inRussian).

Kuzmin, M.I., Khursevich, G.K., Prokopenko, A.A., Fedenya, S.A., Karabanov, E.B.,2009. Centric Diatoms in Lake Baikal during the Late Cenozoic: Morphology,Systematics, Stratigraphy and Stages of Development (Based on the Deep Coresof the Baikal Drilling Project). Academic Publishing House “GEO”, Novosibirsk,p. 374 (in Russian).

Laufer, S., Roy, W., 1972. German Federal Republic Patent 2043629 (Degussa).Lauterbach, S., Brauer, A., Andersen, N., Danielopol, D.L., Dulski, P., Hüls, M.,

Milecka, K., Namiotko, T., Obremska, M., Von Grafenstein, U., Participants, D.,2011. Environmental responses to Lateglacial climatic fluctuations recorded inthe sediments of pre-Alpine Lake Mondsee (northeastern Alps). Journal ofQuaternary Science 26 (3), 253e267.

Last, W.M., Smol, J.P. (Eds.), 2001. Tracking Environmental Change Using LakeSediments. Basin analysis, Coring and Chronological Techniques, vol. 1. KluwerAcademic Publishers, p. 576.

Leng, M.J., Marshall, J.D., 2004. Palaeoclimate interpretation of stable isotope datafrom lake sediment archives. Quaternary Science Reviews 23, 811e831.

Leng, M.J., Barker, P.A., 2006. A review of the oxygen isotope composition oflacustrine diatom silica for palaeoclimate reconstruction. Earth-ScienceReviews 75, 5e27.

Leng, M.J., Wagner, B., Boehm, A., Panagiotopoulos, K., Vane, C.H., Snelling, F.,Haidon, C., Woodley, E., Vogel, H., Zanchetta, G., Baneschi, I., 2013. Under-standing past climatic and hydrological variability in the Mediterranean fromLake Prespa sediment isotope and geochemical record over the Last Glacialcycle. Quaternary Science Reviews 66, 123e136.

Li, L., Mackay, A.W., Zhang, Y., Li, J., 2013. A 1000-year record of vegetation changeand wildfire from maar lake Erlongwan in northeast China. Quaternary Inter-national 290e291, 313e321.

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/

Page 12: The last glacial maximum and late glacial environmental ... · The last glacial maximum and late glacial environmental and climate dynamics in the Baikal region inferred from an oxygen

S.S. Kostrova et al. / Quaternary International xxx (2014) 1e1212

Litt, T., Stebich, M., 1999. Bio- and chronostratigraphy of the lateglacial in the Eifelregion, Germany. Quaternary International 61, 5e16.

Lydolph, P.E., 1977. Climates of the Soviet Union. In: World Survey of Climatology,vol. 7. Elsevier Scientific Publishing Company, Amsterdam-Oxford-New York,p. 443.

Mackay, A.W., 2007. The paleoclimatology of Lake Baikal: a diatom synthesis andprospectus. Earth-Science Reviews 82, 181e215.

Mackay, A.W., Swann, G.E.A., Brewer, T.S., Leng, M.J., Morley, D.W., Piotrowska, N.,Rioual, P., White, D., 2011. A reassessment of late glacial e Holocene diatomoxygen isotope record from Lake Baikal using a geochemical mass-balanceapproach. Journal of Quaternary Science 26 (6), 627e634.

Mackay, A.W., Swann, G.E.A., Fagel, N., Fietz, S., Leng, M.J., Morley, D., Rioual, P.,Tarasov, P., 2013. Hydrological instability during the Last Interglacial in centralAsia: a new diatom oxygen isotope record from Lake Baikal. Quaternary ScienceReviews 66, 45e54.

Meyer, H., Sch€onicke, L., Wand, U., Hubberten, H.-W., Friedrichsen, H., 2000. Isotopestudies of hydrogen and oxygen in ground ice e experiences with the equili-bration technique. Isotopes in Environmental and Health Studies 36, 133e149.

Morley, D.W., Leng, M.J., Mackay, A.W., Sloane, H.J., 2005. Late glacial and Holoceneenvironmental change in the Lake Baikal region documented by oxygen iso-topes from diatom silica. Global and Planetary Change 46, 221e233.

Morley, D.W., Leng, M.J., Mackay, A.W., Sloane, H.J., Rioual, P., Battarbee, R.W., 2004.Cleaning of lake sediment samples for diatom oxygen isotope analysis. Journalof Paleolimnology 31, 391e401.

Moschen, R., Lücke, A., Schleser, G.H., 2005. Sensitivity of biogenic silica oxygenisotopes to changes in surface water temperature and palaeoclimatology.Geophysical Research Letters 32, L07708. http://dx.doi.org/10.1029/2004GL022167.

Moschen, R., Lücke, A., Parplies, J., Radtke, U., Schleser, G.H., 2006. Transfer and earlydiagenesis of biogenic silica oxygen isotope signals during settling and sedi-mentation of diatoms in a temperate freshwater lake (Lake Holzmaar, Ger-many). Geochimica et Cosmochimica Acta 70, 4367e4379.

Müller, S., Tarasov, P.E., Andreev, A.A., Tütken, T., Gartz, S., Diekmann, B., 2010. LateQuaternary vegetation and environments in the Verkhoyansk Mountains region(NE Asia) reconstructed from a 50-ka fossil pollen record from Lake Billyakh.Quaternary Science Reviews 29, 2071e2086.

Müller, S., Tarasov, P.E., Hoelzmann, P., Bezrukova, E.V., Kossler, A., Krivonogov, S.K.,2013. Stable vegetation and environmental conditions during the LastGlacial Maximum: New results from Lake Kotokel (Lake Baikal region, southernSiberia, Russia). Quaternary International. http://dx.doi.org/10.1016/j.quaint.2013.12.012.

Nakagawa, T., Kitagawa, H., Yasuda, Y., Tarasov, P.E., Gotanda, K., Sawai, Y., 2005.Pollen/event stratigraphy of the varved sediment of Lake Suigetsu, central Japanfrom 15,701 to 10,217 SG vyr BP (Suigetsu varve years before present):description, interpretation, and correlation with other regions. QuaternaryScience Reviews 24 (14e15), 1691e1701.

Peck, J.A., Khosbayar, P., Fowell, S.J., Pearce, R.B., Ariunbileg, S., Hansen, B.C.S.,Soninkhishig, N., 2002. Mid to Late Holocene climate change in north centralMongolia as recorded in the sediments of Lake Telmen. Palaeogeography,Palaeoclimatology, Palaeoecology 183, 135e153.

Popovskaya, G.I., Genkal, S.I., Likhoshway, YeV., 2002. Diatoms of the Plankton ofLake Baikal: Atlas and Key. Nauka, Novosibirsk, p. 168 (in Russian).

Prokopenko, A.A., Williams, D.F., Karabanov, E.B., Khursevich, G.K., 2001. Conti-nental response to Heinrich events and Bond cycles in sedimentary record ofLake Baikal, Siberia. Global and Planetary Change 28, 217e226.

Prokopenko, A.A., Williams, D.F., 2004. Deglacial methane emission signals in thecarbon isotopic record of Lake Baikal. Earth and Planetary Science Letters 218,135e147.

Prokopenko, A.A., Kuzmin, M.I., Williams, D.F., Gelety, V.F., Kalmychkov, G.V.,Gvozdkov, A.N., Solotchin, P.A., 2005. Basin-wide sedimentation changes anddeglacial lake-level rise in the Hovsgol basin, NW Mongolia. Quaternary In-ternational 136, 59e69.

Ryves, D.V., Jewson, D.H., Sturm, M., Battarbee, R.W., Flower, R.J., Mackay, A.W.,Granin, N.G., 2003. Quantitative and qualitative relationships between plank-tonic diatom communities and diatom assemblages in sedimenting materialand surface sediments in Lake Baikal, Siberia. Limnology and Oceanography 48,1643e1661.

Rodrigues, T., Grimalt, J.O., Abrantes, F., Naughton, F., Flores, J., 2010. The last gla-cialeinterglacial transition (LGIT) in the western mid-latitudes of the NorthAtlantic: abrupt sea surface temperature change and sea level implications.Quaternary Science Reviews 29, 1853e1862.

Rozanski, K., Aragu�as-Aragu�as, L., Gonfiantini, R., 1993. Isotopic patterns in modernglobal precipitation. Geophysical Monograph. In: Climate Change in ContinentalIsotope Records, vol. 78. American Geophysical Union, pp. 1e36.

Please cite this article in press as: Kostrova, S.S., et al., The last glacial mBaikal region inferred from an oxygen isotope record of lacustrine diatoj.quaint.2014.07.034

Rühland, K.M., Paterson, A.M., Smol, J.P., Pienitz, R., 2008. Hemispheric-scale pat-terns of climate-related shifts in planktonic diatoms from North American andEuropean lakes. Global Change Biology 14, 2740e2754.

Seal, R.R., Shanks, W.C., 1998. Oxygen and hydrogen isotope systematics of LakeBaikal, Siberia: implications for paleoclimate studies. Limnology and Ocean-ography 43, 1251e1261.

Shichi, K., Takahara, H., Krivonogov, S.K., Bezrukova, E.V., Kashiwaya, K.,Takehara, A., Nakamura, T., 2009. Late Pleistocene and Holocene vegetation andclimate records from Lake Kotokel, central Baikal region. Quaternary Interna-tional 205 (1e2), 98e110.

Shichi, K., Takahara, H., Hase, Y., Watanabe, T., Nara, F.W., Nakamura, T., Tani, Y.,Kawai, T., 2013. Vegetation response in the southern Lake Baikal region toabrupt climate events over the past 33 cal kyr. Palaeogeography, Palae-oclimatology, Palaeoecology 375, 70e82.

Sirocko, F. (Ed.), 2012. Wetter, Klima, Menschheitsentwicklung von der Eiszeit bisins 21. Jahrhundert. Konrad Theiss Verlag, Darmstadt, p. 208.

Sklyarov, E.V., Solotchina, E.P., Vologina, E.G., Ignatova, N.V., Izokh, O.P.,Kulagina, N.V., Sklyarova, O.A., Solotchin, P.A., Stolpovskaya, V.N., Ukhova, N.N.,Fedorovskii, V.S., Khlystov, O.M., 2010. Detailed Holocene climate record fromthe carbonate section of saline Lake Tsagan-Tyrm (West Baikal area). RussianGeology and Geophysics 51, 237e258.

Smol, J.P., 1988. Palaeoclimate proxy data from freshwater diatoms. Verhandlungendes Internationalen Verein Limnologie 23, 837e844.

Stebich, M., Mingram, J., Han, J., Liu, J., 2009. Late Pleistocene spread of (cool-)temperate forests in Northeast China and climate changes synchronous withthe North Atlantic region. Global and Planetary Change 65, 56e70.

Subetto, D.A., 2009. Donnie otlozheniya ozer: paleolimnologicheskie reconstruktsii.Izdatel‘tsvo RGPY im. A.I. Gertsena, Moscow, p. 339 (in Russian).

Svensson, A., Andersen, K.K., Bigler, M., Clausen, H.B., Dahl-Jensen, D., Davies, S.M.,Johnsen, S.J., Muscheler, R., Parrenin, F., Rasmussen, S.O., R€othlisberger, R.,Seierstad, I., Steffensen, J.P., Vinther, B.M., 2008. A 60000 year Greenlandstratigraphic ice core chronology. Climate of the Past 4, 47e57.

Swann, G.E.A., Maslin, M.A., Leng, M.J., Sloane, H.J., Haug, G.H., 2006. Diatom d18Oevidence for the development of the modern halocline system in the subarcticnorth-west Pacific at the onset of major Northern Hemisphere glaciation.Paleoceanography 21, PA1009. http://dx.doi.org/10.1029/2005PA001147.

Swann, G.E.A., Leng, M.J., Sloane, H.J., Maslin, M.A., 2008. Isotope offsets in marinediatom d18O over the last 200 ka. Journal of Quaternary Science 23, 389e400.

Swann, G.E.A., Leng, M.J., 2009. A review of diatom d18O in palaeoceanography.Quaternary Science Reviews 28, 384e398.

Swann, G.E.A., Leng, M.J., Juschus, O., Melles, M., Brigham-Grette, J., Sloane, H.J.,2010. A combined oxygen and silicon diatom isotope record of Late Quaternarychange in Lake El’gygytgyn, North East Siberia. Quaternary Science Reviews 29,774e786.

Tarasov, P., Bezrukova, E., Karabanov, E., Nakagava, T., Wagner, M., Kulagina, N.,Letunova, P., Abzaeva, A., Granoszewski, W., Riedel, F., 2007. Vegetation andclimate dynamics during the Holocene and Eemian interglacials derived fromLake Baikal pollen records. Palaeogeography, Palaeoclimatology, Palaeoecology252, 440e457.

Tarasov, P.E., Bezrukova, E.V., Krivonogov, S.K., 2009. Late Glacial and Holocenechanges in vegetation cover and climate in southern Siberia derived from a 15kyr long pollen record from Lake Kotokel. Climate of the Past 5, 285e295.

Wang, Y.J., Cheng, H., Edwards, R.L., An, Z.S., Wu, J.Y., Shen, C.-C., Dorale, J.A., 2001.A high-resolution absolute-dated Late Pleistocene Monsoon record from HuluCave, China. Science 294, 2345e2348.

Wang, L., Li, J., Lu, H., Gu, Z., Rioual, P., Hao, Q., Mackay, A.W., Jiang, W., Cai, B., Xu, B.,Han, J., Chu, G., 2012a. The East Asian winter monsoon over the last 15,500years: its links to high-latitudes and tropical climate systems and complexcorrelation to the summer monsoon. Quaternary Science Reviews 32, 131e142.

Wang, Q., Yang, X., Hamilton, P.B., Zhang, E., 2012b. Linking spatial distributions ofsediment diatom assemblages with hydrological depth profiles in a plateaudeepewater lake system of subtropical China. Fottea 12 (1), 59e73.

Yu, Z., Eicher, U., 2001. Three amphi-Atlantic century-scale cold events during theBølling e Allerød warm period. Geographie physique et Quaternaire 55 (2),171e179.

Yuan, D.X., Cheng, H., Edwards, R.L., Dykoski, C.A., Kelly, M.J., Zhang, M.L., Qing, J.M.,Lin, Y.S., Wang, Y.J., Wu, J.Y., Dorale, J.A., An, Z.S., Cai, Y.J., 2004. Timing, duration,and transitions of the Last Interglacial Asian monsoon. Science 304, 575e578.

Zhang, Y., Wünnemann, B., Bezrukova, E.V., Ivanov, E.V., Shchetnikov, A.A.,Nourgaliev, D., Levina, O.V., 2013. Basin morphology and seismic stratigraphy ofLake Kotokel, Baikal region, Russia. Quaternary International 290e291, 57e67.

Zhao, K., Li, X., Dodson, J., Zhou, X., Atahan, P., 2013. Climate instability during thelast deglaciation in central Asia, reconstructed by pollen data from Yili Valley,NW China. Review of Palaeobotany and Palynology 189, 8e17.

aximum and late glacial environmental and climate dynamics in them silica, Quaternary International (2014), http://dx.doi.org/10.1016/